Skip to main content


Fig. 16 | BMC Bioinformatics

Fig. 16

From: SVM-RFE: selection and visualization of the most relevant features through non-linear kernels

Fig. 16

C statistics results by method and ranked variable in the DLBCL dataset. The X-axis shows the rank of each one of the variables in the dataset after applying the RFE algorithm. The lower the rank the more relevant the variable is and the larger the C statistic is expected. As each method can rank differently the variables, given a rank the variable can be different between methods, due to this the C statistic (Y-axis) is different

Back to article page